Neuromorphic Wireless Cognition: Event-Driven Semantic Communications for Remote Inference
Neuromorphic Wireless Cognition: Event-Driven Semantic Communications for Remote Inference
复制标题
DOI：
10.1109/tccn.2023.3236940
复制
发表时间：
2022-06
影响因子：
8.6
通讯作者：
Jiechen Chen;N. Skatchkovsky;O. Simeone
中科院分区：
文献类型：
--
作者：
Jiechen Chen;N. Skatchkovsky;O. Simeone
神经形态计算是一种新兴的计算范式，它从批量处理转向在线、事件驱动的流数据处理。神经形态芯片，当与基于尖峰的传感器相结合时，可以通过仅在尖峰定时记录相关事件时消耗能量以及通过对环境中变化的条件提供低延迟响应来固有地适应数据分布的“语义”。本文提出了一种端到端的设计，神经形态的无线物联网系统，集成了基于尖峰的传感，处理和通信。在建议的NeuroComm系统中，每个传感设备都配备了一个神经形态传感器，尖峰神经网络（SNN），和脉冲无线电（IR）发射机与多个天线。在共享衰落信道上向配备有多天线脉冲无线电接收机和SNN的接收机进行传输。为了使接收机适应衰落信道条件，我们引入了一个超网络来控制使用导频的解码SNN的权重。导频、编码SNN、解码SNN和超网络在多个信道实现中联合训练。所提出的系统显着改善传统的基于帧的数字解决方案，以及替代非自适应训练方法，在时间精度和能耗指标。
Neuromorphic computing is an emerging computing paradigm that moves away from batched processing towards the online, event-driven, processing of streaming data. Neuromorphic chips, when coupled with spike-based sensors, can inherently adapt to the “semantics” of the data distribution by consuming energy only when relevant events are recorded in the timing of spikes and by proving a low-latency response to changing conditions in the environment. This paper proposes an end-to-end design for a neuromorphic wireless Internet-of-Things system that integrates spike-based sensing, processing, and communication. In the proposed NeuroComm system, each sensing device is equipped with a neuromorphic sensor, a spiking neural network (SNN), and an impulse radio (IR) transmitter with multiple antennas. Transmission takes place over a shared fading channel to a receiver equipped with a multi-antenna impulse radio receiver and with an SNN. In order to enable adaptation of the receiver to the fading channel conditions, we introduce a hypernetwork to control the weights of the decoding SNN using pilots. Pilots, encoding SNNs, decoding SNN, and hypernetwork are jointly trained across multiple channel realizations. The proposed system is shown to significantly improve over conventional frame-based digital solutions, as well as over alternative non-adaptive training methods, in terms of time-to-accuracy and energy consumption metrics.